You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@iotdb.apache.org by "Haimei Guo (Jira)" <ji...@apache.org> on 2021/01/26 09:16:00 UTC
[jira] [Commented] (IOTDB-1128) 按照自然月份的降频聚合查询，结果集中的起始时间不正确

    [ https://issues.apache.org/jira/browse/IOTDB-1128?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17271989#comment-17271989 ] 

Haimei Guo commented on IOTDB-1128:
-----------------------------------

当前版本月聚合是按照30天划分时间间隔的。新的版本，月聚合会改为按照自然月份聚合。

自然月聚合的pr 链接：https://github.com/apache/iotdb/pull/2029

> 按照自然月份的降频聚合查询，结果集中的起始时间不正确
> --------------------------
>
>                 Key: IOTDB-1128
>                 URL: https://issues.apache.org/jira/browse/IOTDB-1128
>             Project: Apache IoTDB
>          Issue Type: Bug
>         Environment: 0.11.3-SNAPSHOT
>            Reporter: 刘珍
>            Priority: Minor
>
> 测试用例，有问题的sql在最后1条：
> create timeseries root.db_1.tab1.temp with datatype=INT64,encoding=REGULAR ;
> insert into root.db_1.tab1(time,temp) values(2017-11-01T00:00:00,110101);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T01:00:00,110102);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T02:00:00,110103);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T03:00:00,110104);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T04:00:00,110105);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T05:00:00,110106);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T00:00:00,110101);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T01:00:00,110102);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T02:00:00,110103);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T03:00:00,110104);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T04:00:00,110105);
> insert into root.db_1.tab1(time,temp) values(2017-11-01T05:00:00,110106);
> insert into root.db_1.tab1(time,temp) values(2017-11-02T00:00:00,110201);
> insert into root.db_1.tab1(time,temp) values(2017-11-02T01:00:00,110202);
> insert into root.db_1.tab1(time,temp) values(2017-11-02T02:00:00,110203);
> insert into root.db_1.tab1(time,temp) values(2017-11-02T03:00:00,110204);
> insert into root.db_1.tab1(time,temp) values(2017-11-02T04:00:00,110205);
> insert into root.db_1.tab1(time,temp) values(2017-11-02T05:00:00,110206);
> insert into root.db_1.tab1(time,temp) values(2017-11-03T00:00:00,110301);
> insert into root.db_1.tab1(time,temp) values(2017-11-03T01:00:00,110302);
> insert into root.db_1.tab1(time,temp) values(2017-11-03T02:00:00,110303);
> insert into root.db_1.tab1(time,temp) values(2017-11-03T03:00:00,110304);
> insert into root.db_1.tab1(time,temp) values(2017-11-03T04:00:00,110305);
> insert into root.db_1.tab1(time,temp) values(2017-11-03T05:00:00,110306);
> insert into root.db_1.tab1(time,temp) values(2017-11-04T00:00:00,110401);
> insert into root.db_1.tab1(time,temp) values(2017-11-04T01:00:00,110402);
> insert into root.db_1.tab1(time,temp) values(2017-11-04T02:00:00,110403);
> insert into root.db_1.tab1(time,temp) values(2017-11-04T03:00:00,110404);
> insert into root.db_1.tab1(time,temp) values(2017-11-04T04:00:00,110405);
> insert into root.db_1.tab1(time,temp) values(2017-11-04T05:00:00,110406);
> select count(temp), max_value(temp) from root.db_1.tab1 group by ([2017-11-01T00:00:00, 2017-11-05T03:18:10.000+08:00),1d);
> select count(temp), max_value(temp) from root.db_1.tab1 group by ([2017-11-01T00:00:00, 2017-11-05T03:18:10.000+08:00),3h,1d);
> 按照自然月份的降频聚合查询
> insert into root.db_1.tab1(time,temp) values(2017-12-04T00:00:00,110401);
> insert into root.db_1.tab1(time,temp) values(2017-12-04T01:00:00,110402);
> insert into root.db_1.tab1(time,temp) values(2017-12-04T02:00:00,110403);
> insert into root.db_1.tab1(time,temp) values(2017-12-04T03:00:00,110404);
> insert into root.db_1.tab1(time,temp) values(2017-12-04T04:00:00,110405);
> insert into root.db_1.tab1(time,temp) values(2017-12-04T05:00:00,110406);
> insert into root.db_1.tab1(time,temp) values(2018-01-04T00:00:00,110401);
> insert into root.db_1.tab1(time,temp) values(2018-01-04T01:00:00,110402);
> insert into root.db_1.tab1(time,temp) values(2018-01-04T02:00:00,110403);
> insert into root.db_1.tab1(time,temp) values(2018-01-04T03:00:00,110404);
> insert into root.db_1.tab1(time,temp) values(2018-01-04T04:00:00,110405);
> insert into root.db_1.tab1(time,temp) values(2018-01-04T05:00:00,110406);
> insert into root.db_1.tab1(time,temp) values(2018-01-04T06:00:00,110407);
> insert into root.db_1.tab1(time,temp) values(2018-02-04T00:00:00,110401);
> insert into root.db_1.tab1(time,temp) values(2018-02-04T01:00:00,110402);
> insert into root.db_1.tab1(time,temp) values(2018-02-04T02:00:00,110403);
> insert into root.db_1.tab1(time,temp) values(2018-02-04T03:00:00,110404);
> insert into root.db_1.tab1(time,temp) values(2018-02-04T04:00:00,110405);
> insert into root.db_1.tab1(time,temp) values(2018-02-04T05:00:00,110406);
> select count(temp), max_value(temp) from root.db_1.tab1 where time > 2017-11-01T00:00:00 group by ([2017-11-01T00:00:00,2018-02-04T05:00:00),1mo, 2mo);
> +-----------------------------+--------------------------+------------------------------+
> | Time|count(root.db_1.tab1.temp)|max_value(root.db_1.tab1.temp)|
> +-----------------------------+--------------------------+------------------------------+
> |2017-11-01T00:00:00.000+08:00| 23| 110406|
> |2017-12-31T00:00:00.000+08:00| 7| 110407|
> +-----------------------------+--------------------------+------------------------------+
> Total line number = 2
> 2017-12-31T00:00:00.000+08:00这个起始时间不正确，应该是2018-01-01



--
This message was sent by Atlassian Jira
(v8.3.4#803005)